Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)T

Thorry

@ Thorry @feddit.org

Posts
2
Comments
302
Joined
1 yr. ago

  • What is with the pro-AI posts lately?

    Jump

  • Out of curiosity, why have you been expecting a softening? From what I've seen AI tools for coding have gotten worse recently, not better. And companies are now jacking up the prices, to be more in line with costs. I've heard people irl complaining they went from $10 per month to $1000 if they were to continue using it the same way. Most have capped themselves or stopped altogether, as with that price it isn't worth it anymore.

    So my personal experience is more people complaining, but I'm interested in your view.

  • Surely that gets Geordi off the hook.

    Jump

  • What color is the letter E?

    Jump

  • To me it's more like yellow.

  • The Centaur and the Mermaid [Pepper & Carrot]

    Jump

  • The intro theme is a banger tho

  • The AI operator: Biggest role in Silicon Valley

    Jump

  • Suuuuure, sloperators are going to be big.

    Maybe let the tech speak for itself instead of forcing all this slop down our throats. The harder people shout "THIS IS THE NEXT BIG THING!" the more I don't believe it. All of these comparisons to old tech that became big, but don't look at the thousands of ideas that never became big. Just because some tech got big, does not mean any tech will be big and the odds are very much against any single tech becoming big.

    And these comparisons to the dotcom bubble, sure it was all a bubble, but the tech was valid! That's true, but we didn't know that at the time. And the bubble popping still sucked, the economy was bad, people lost their jobs, regular folk were suffering for VC going ham on some new tech. And it also turns cause and effect on its head. It wasn't like the dotcom tech needed the bubble to become successful or that tech can't become successful without a bubble. Just because AI is now in a big bubble doesn't mean anything useful will come out the other end.

    Personally I think the future of AI is more in small dedicated expert systems, that do select specific roles and do them well. Not these know all, do all, everything chat based LLM systems they are pushing now. And I believe we will have specific chips that run these systems in an efficient local way, not the subscription based future tech is pushing now. I also believe there is a hard ceiling to this tech, where it can be pushed so far and not further and I feel we are close to that ceiling now. Going beyond requires exponential more time and effort to setup and run and in the end there just isn't a use-case available where it's cost effective to do so.

    But hey, I'm just an idiot on the internet, what do I know.

  • Turned my old headphones into wireless headphones

    Jump

  • Yeah I had the exact same experience. On the other hand, my $16 wireless earbuds have excellent audio quality. So it seems there are good cheap DACs out there, but also a lot of trash.

  • Superb cool

    Jump

  • And keep in mind there weren't just large beasts, but small ones as well. People tend to focus on the big ones, because they are most impressive, but they came in the full range of sizes. Just imagine how alien our planet looked compared to what we are used to now.

  • AI finds signs of pancreatic cancer before tumors develop

    Jump

  • Agreed, this is exactly what reinforcement learning and neural networks are good at. Calling them AI is beyond dumb, but hey marketing will be marketing. It's pattern recognition, which is cool, but nobody would call that intelligent otherwise. Another big issue with the marketing is they only report on the success rate and not the failure rate. Doctors praise the cases being caught, but dislike the models pointing out stuff that is clearly not a tumor. It wastes time for people already short on time. These models also risk doctors becoming over reliant on them, even though they can have serious blind spots and thus miss stuff a doctor would have caught. Or the other way around, have people receive treatment (often not without risk, discomfort and cost to the patient), where none was needed. The thing that bothers me the most is how it's always framed as a win for AI. Like see AI is good at diagnosing cancer (which then gets extrapolated to curing cancer for some bizarre reason), so that useless chat bot is also good somehow. Because AI.

  • What do you call an egg that knows martial arts? An egg-spert fighter!

    Jump

  • And Mr. Neelix was never seen ever again.

  • Randy Bobandy

    Jump

  • FUCK OFF RANDY

  • How long is coffee good for in the fridge?

    Jump
  • Deleted

    Permanently Deleted

    Jump

  • To be fair, there is a whole lot more to selling software than just making it. There's this myth out there if you just build the right software at the right time, success is guaranteed. But in reality all of the success stories from the past had many competitors, often with better made software. There's many factors that go into it, luck is a big factor and upfront cash is also a big one (especially nepotism is a common theme).

  • Market slumps as OpenAI reportedly misses internal targets for active users and revenue — Nvidia, Oracle, AMD, and CoreWeave shares all tremble on the news

    Jump

  • I do hope this bubble pops, but this kind of reporting is pretty useless. It's like all those Tesla stock posts we got last year. They report every time the stock goes down, even when the overall trend is the stock going up.

    Nvidia is still up 28% from last month. Amd is up almost 60%, Oracle is up 20% and CoreWeave is up 41%.

    When looking at 1Y these stocks are up as well. These stocks have gotten a big boost this month, so a correction isn't that big of a deal.

    The markets are absolutely insane at the moment. We are in the middle of a huge energy crisis, one that hasn't even shown its true face. Yet somehow the stocks are up. There's more unrest than ever before, the spending power is down, the trust in the economy and future is down. Yet somehow stocks are up. All this AI nonsense is at spending levels it's just a straight up money pit, even the best projections for revenue (let alone profit) can't make back the money spent. Yet somehow stocks are up. All of these circlejerk investments and ghost datacenters scams are going on. Yet somehow stocks are up.

    The stock market has fundamentally broken and isn't really an indicator of anything anymore. To paraphrase an idiot: Stock goes up, stock goes down, you can't explain that.

  • Don't pay for AI, frame your questions like you want Maccas.

    Jump

  • I think McDonald's UK still has a support bot. But it's like one of those pre-LLM bots that does very basic stuff. Basically a glorified search function for the website.

  • Removed

    Reminder: Reminder: This is STILL NOT a Politics community

    Jump

  • Be the change you want to see in the world

  • Deleted

    Permanently Deleted

    Jump

  • While it's true that the big players don't use consumer grade hardware, there is an actual shortage of consumer grade stuff. This is pushing prices of whole computers, laptops, phones and hardware such as gpus and ram sticks to high prices.

    The reason is there's only a few companies creating high end memory, storage and processing chips like cpus and gpus. These companies have in the past 2 years shifted their manufacturing capabilities to creating just the high margin enterprise stuff. This at the cost of producing the lower end low margin consumer stuff. For a while we were coasting on existing stock, but once that began to ran out, prices started going up real fast.

    Big players like Dell, Lenovo and HP for example bought up a whole lot of hardware so they could have a 2026 lineup of products. This again further pushed up prices for consumer hardware, especially memory. And even these companies are struggling with getting enough supplies for a 2027 range. So expect prices to go up further the next couple of months. Although in the consumer space we are seeing prices stabilize, because demand has fallen off hard, not many people can or are willing to afford as much as the asking price.

    I agree totally it's fucking bullshit and all of the AI speculation impacting the rest of the markets so hard is a real issue, the shortages are real and not just market speculation.

  • What is your favourite moon(s) in the Solar System?

    Jump

  • Are you thinking of "The moon is a harsh mistress" by Heinlein? That has a theme about growing a lot of wheat on the Moon and exporting it to Earth.

    I read that book a long time ago and liked it, even though it had so many flaws.

  • What is your favourite moon(s) in the Solar System?

    Jump

  • I'm gonna go with Hyperion. My man is chaos in every way possible. Very unpredictable, yet somehow stable. Not tidally locked to Saturn like a bitch, just doing it's own thing. Not even round, despite being rather on the chunky side. Some might say ugly, others would say interesting. Not very dense either.

    I love how much this moon is like me.

  • Dr Douchebag to med bay

    Jump

  • I think it's really hard to get a picture of Terry and have it not be super hot.